The objectives of this application are to determine the key environmental and genetic factors that lead to symptomatic recurrence among stone formers, as well as to validate asymptomatic stone formation and growth by CT scan as a surrogate for symptomatic recurrence. The central clinical hypothesis is that kidney stone recurrence can be predicted from clinical, laboratory, and radiographic measurements in the electronic health record (EHR). Our central mechanistic hypothesis is that genomic markers for kidney stones contribute to an increased risk of incident and recurrent stone events. In order to develop effective prevention strategies for kidney stone recurrence, we plan to objectively test our hypotheses with the following aims: Specific Aim #1: To develop a model to better predict symptomatic stone recurrence using clinical and laboratory information from the comprehensive (inpatient and outpatient) health records of 4680 chart validated symptomatic stone formers in the Olmsted County general population (1984 to 2016). Specific Aim #2: To determine if urine chemistries, blood serologies, and life-style factors can improve the prediction of symptomatic recurrence beyond the clinical and laboratory characteristics available in the EHR using 800 incident stone formers in our expanded prospective cohort (2009 to 2017). Specific Aim #3: To determine if models predicting symptomatic recurrence in Aims 1 or 2 also predict radiographic stone formation and growth among 300 incident stone formers in our current prospective cohort. Specific Aim #4: To sample stone formers from the general population to identify causative genes. A) Assemble a cohort of 1500 carefully validated and phenotyped incident kidney stone formers, plus 1500 controls for genetic analysis. Multiplex families will be a focus of this recruitment effort. B) Screen 400 validated incident symptomatic kidney stone formers for significant variants in 66 candidate genes for hypercalcuria, employing exon capture and next generation sequencing.